IFRS 9 and CECL Credit Risk Modelling and Validation

IFRS 9 and CECL Credit Risk Modelling and Validation

A Practical Guide with Examples Worked in R and SAS

1st Edition - January 15, 2019
This is the Latest Edition
  • Author: Tiziano Bellini
  • eBook ISBN: 9780128149416
  • Paperback ISBN: 9780128149409

Purchase options

Purchase options
DRM-free (Mobi, EPub, PDF)
Sales tax will be calculated at check-out

Institutional Subscription

Free Global Shipping
No minimum order


IFRS 9 and CECL Credit Risk Modelling and Validation covers a hot topic in risk management. Both IFRS 9 and CECL accounting standards require Banks to adopt a new perspective in assessing Expected Credit Losses. The book explores a wide range of models and corresponding validation procedures. The most traditional regression analyses pave the way to more innovative methods like machine learning, survival analysis, and competing risk modelling. Special attention is then devoted to scarce data and low default portfolios. A practical approach inspires the learning journey. In each section the theoretical dissertation is accompanied by Examples and Case Studies worked in R and SAS, the most widely used software packages used by practitioners in Credit Risk Management.

Key Features

  • Offers a broad survey that explains which models work best for mortgage, small business, cards, commercial real estate, commercial loans and other credit products
  • Concentrates on specific aspects of the modelling process by focusing on lifetime estimates
  • Provides an hands-on approach to enable readers to perform model development, validation and audit of credit risk models


Upper-division undergraduates, graduate students, and professionals working in economic modelling and statistics.

Table of Contents

  •  1. Introduction to Expected Credit Loss Modelling and Validation
    1.1 Introduction         
    1.2 IFRS 9 
    1.21 Staging Allocation      
    1.22 ECL Ingredients
    1.23 Scenario Analysis and ECL
    1.3 CECL 
    1.31 Loss-Rate Methods    
    1.32 Vintage Methods
    1.33 Discounted Cash Flow Methods
    1.34 Probability of Default Method (PD, LGD, EAD)
    1.35 IFRS 9 vs CECL 
    1.4 ECL and Capital Requirements
    1.41 Internal Rating-Based Credit Risk-Weighted Assets
    1.42 How ECL Affects Regulatory Capital and Ratios
    1.5 Book Structure at a Glance
    1.6 Summary          

    2. One-Year PDs
    2.1 Introduction                  
    2.2 Default Definition and Data Preparation 
    2.21 Default Definition   
    2.22 Data Preparation 
    2.3 Generalized Linear Models (GLMs) 
    2.31 GLM (Scorecard) Development
    2.32 GLM Calibration
    2.33 GLM Validation
    2.4 Machine Learning (ML) Modelling
    2.41 Classification and Regression Trees (CART)
    2.42 Bagging, Random Forest, and Boosting
    2.43 ML Model Calibration
    2.44 ML Model Validation
    2.5 Low Default Portfolio, Market-Based, and Scarce Data Modelling
    2.51 Low Default Portfolio Modelling
    2.52 Market Based Modelling
    2.53 Scarce Data Modelling
    2.54 Hints on Low Default Portfolio, Market-Based, and Scarce Data Model Validation 
    2.6 SAS Laboratory           
    2.7 Summary               
    2.8 Appendix A From Linear Regression to GLMs
    2.9 Appendix B Discriminatory Power Assessment

    3. Lifetime PDs 1
    3.1 Introduction
    3.2 Data Preparation      
    3.21 Default Flag Creation 
    3.22 Account-Level (Panel) Database Structure
    3.3 Lifetime GLM Framework
    3.31 Portfolio-level GLM Analysis
    3.32 Account-Level GLM Analysis
    3.33 Lifetime GLM Validation
    3.4 Survival Modelling 
    3.41 Kaplan Meier (KM) Survival Analysis
    3.42 Cox Proportional Hazard (CPH) Survival Analysis
    3.43 Accelerated Failure Time (AFT) Survival Analysis
    3.44 Survival Model Validation
    3.5 Lifetime Machine Learning (ML) Modelling
    3.51 Bagging, Random Forest, and Boosting Lifetime PD
    3.52 Random Survival Forest Lifetime PD
    3.53 Lifetime ML Validation
    3.6 Transition Matrix Modelling
    3.61 Na_ve Markov Chain Modelling
    3.62 Merton-Like Transition Modelling
    3.63 Multi State Modelling
    3.64 Transition Matrix Model Validation
    3.7 SAS Laboratory 
    3.8 Summary  

    4. LGD Modelling
    4.1 Introduction
    4.2 LGD Data Preparation
    4.21 LGD Data Conceptual Characteristics 
    4.22 LGD Database Elements
    4.3 LGD Micro-Structure Approach
    4.31 Probability of Cure       
    4.32 Severity
    4.33 Defaulted Asset LGD
    4.34 Forward-Looking Micro-Structure LGD Modelling
    4.35 Micro-Structure Real Estate LGD Modelling
    4.36 Micro-Structure LGD Validation
    4.4 LGD Regression Methods
    441 Tobit Regression
    4.42 Beta Regression
    4.43 Mixture Models and forward-looking Regression
    4.44 Regression LGD Validation
    4.5 LGD Machine Learning (ML) Modelling
    4.51 Regression Tree LGD
    4.52 Bagging, Random Forest, and Boosting LGD
    4.53 Forward-Looking Machine Learning LGD
    4.54 Machine Learning LGD Validation
    4.6 Hints on LGD Survival Analysis
    4.7 Scarce Data and Low Default Portfolio LGD Modelling
    4.71 Expert Judgement LGD Process
    4.72 Low Default Portfolio LGD
    4.73 Hints on How to Validate Scarce Data and Low Default Portfolio LGDs
    4.8 SAS Laboratory
    4.9 Summary

    5. Prepayments, Competing Risks and EAD Modelling
    5.1 Introduction
    5.2 Data Preparation
    5.21 How to Organize Data
    5.3 Full Prepayment Modelling
    5.31 Full Prepayment via GLMs
    5.32 Machine Learning (ML) Full Prepayment Modelling
    5.33 Hints on Survival Analysis
    5.34 Full Prepayment Model Validation
    5.4 Competing Risk Modelling
    5.41 Multinomial Regression Competing Risks Modelling
    5.42 Full Evaluation Procedure
    5.43 Competing Risk Model Validation
    5.5 EAD Modelling
    5.51 A Competing-Risk-Like EAD Framework
    5.52 Hints on EAD Estimation via Machine Learning (ML)
    5.53 EAD Model Validation
    5.6 SAS Laboratory      
    5.7 Summary

    6. Scenario Analysis and Expected Credit Losses
    6.1 Introduction
    6.2 Scenario Analysis
    6.21 Vector Auto-Regression (VAR) and Vector Error-Correction (VEC) Modelling
    6.22 VAR and VEC Forecast
    6.23 Hints on GVAR Modelling
    6.3 ECL Computation in Practice 
    6.31 Scenario Design and Satellite Models
    6.32 Lifetime ECL
    6.33 IFRS 9 Staging Allocation
    6.4 ECL Validation
    6.41 Historical and Forward-Looking Validation
    6.42 Credit Portfolio Modelling and ECL Estimation
    6.5 SAS Laboratory
    6.6 Summary

Product details

  • No. of pages: 316
  • Language: English
  • Copyright: © Academic Press 2019
  • Published: January 15, 2019
  • Imprint: Academic Press
  • eBook ISBN: 9780128149416
  • Paperback ISBN: 9780128149409
  • About the Author

    Tiziano Bellini

    Tiziano Bellini received his PhD degree in statistics from the University of Milan after being a visiting PhD student at the London School of Economics and Political Science. He is Qualified Chartered Accountant and Registered Auditor. He gained wide risk management experience across Europe, in London, and in New York. He is currently Director at BlackRock Financial Market Advisory (FMA) in London. Previously he worked at Barclays Investment Bank, EY Financial Advisory Services in London, HSBCs headquarters, Prometeia in Bologna, and other leading Italian companies. He is a guest lecturer at Imperial College in London, and at the London School of Economics and Political Science. Formerly, he served as a lecturer at the University of Bologna and the University of Parma. Tiziano is author of Stress Testing and Risk Integration in Banks, A Statistical Framework and Practical Software Guide (in Matlab and R) edited by Academic Press. He has published in the European Journal of Operational Research, Computational Statistics and Data Analysis, and other top-reviewed journals. He has given numerous training courses, seminars, and conference presentations on statistics, risk management, and quantitative methods in Europe, Asia, and Africa.

    Affiliations and Expertise

    BlackRock Financial Market Advisory, London, UK

    Latest reviews

    (Total rating for all reviews)

    • HUSEYINGONUL Sat Dec 19 2020

      Great Book

      The book has a very good coverage on the matter with examples and insights.

    • ElderNunes Mon Sep 16 2019

      Very clear book

      All concepts very well explained in tables which facilitates comprehension. Highly recommended

    • Sulagn Thu May 16 2019

      Great Book for practitioners. Any idea how to access the dataset used in the examples

      Great Read!

    • Julian O. Wed Mar 13 2019

      Ifrs 9 and cecl credit risk modelling and validation

      Perfect for risk amateurs

    • Matej Thu Feb 28 2019

      Overall great, but not without some shortcomings

      As the majority of banks use standardized approach, I would love to have some more discussion about the comparison of parameter calculation across approaches, e.g. of EAD. You can't know how EAD is calculated under the standardized approach from this book, also CCF table could be provided and an example shown. Same goes for a LGD calculation and haircuts. Sure, professionals will know, but students without experience won't. Furthermore, author doesn't provide a comprehensive calculation example of ECL across multiple years, where marginal PDs, lifetime PD, and marginal ECLs would be shown, because there are always some simplifications. So a little more discussions about a PD term structure, how it is affected by economic cycle, as well as providing examples of bad practices would be of added value: for example a few pages on when probability-weighted economic scenarios is not unbiased.